The long-term objective of this research program is to understand how behavioral factors such as physical activity can modify risk for cardiovascular disease, including diabetes. Type 2 diabetes has been increasing in children, suggesting that prevention efforts are needed during childhood. Insulin resistance (i.e., low insulin sensitivity), a central feature of type 2 diabetes, is associated with sedentary behavior, poor aerobic fitness, general and visceral adiposity, hypertension, inflammation, hemostatic dysfunction, dyslipidemia, African-American ethnicity, and arteriosclerosis risk. Increased physical activity results in reduced risk of diabetes and cardiovascular disease in adults. No experimental studies have examined dose-response relationships of exercise with insulin sensitivity and related health indicators in children. This application is a randomized, controlled dose-response trial to determine the effect of 4 mo of intensive physical training at two dose levels (40 min aerobic exercise, 5d/wk vs. 20 min, 5d/wk) on insulin sensitivity (via oral glucose tolerance test) in 240 overweight, sedentary African-American and white boys and girls in third grade. The specific aims of the study test the hypotheses that the high-dose group will improve more than the low-dose group, which is expected to improve relative to controls. Improvements in body composition (% body fat, visceral fat) and aerobic fitness will be tested to see whether they can account for improvements in insulin sensitivity. Additional aims of the study will explore the effects of exercise dose on factors associated with insulin resistance and cardiovascular risk (blood pressure, lipid/lipoprotein measures, C-reactive protein, fibrinogen), and whether the change in insulin sensitivity can account for any such improvements. Ethnic and sex differences in response to intervention will be assessed. It is possible that the stimulation of insulin sensitivity takes place within the first 20 min of exercise each day and greater amounts of exercise are unnecessary (a threshold effect). The youths assigned to the longer-duration sessions may have difficulty maintaining the high-intensity exercise, with the result that they do not actually engage in a greater volume of exercise per session. By monitoring heart rate, we will be able to see if this occurs. Since so little is known in children about this important aspect of exercise dose, this project promises to contribute substantially to our understanding of how much exercise to recommend to obese youths in order to enhance their fitness, body composition and health comparison of effects of these different volumes of exercise would provide evidence for best practice guidelines and assist in designing large-scale health promotion programs for children.